Iranian License Plate Character Recognition Using Classification Ensemble

Vehicle license plate recognition system is an important research topic of the applications of intelligent transportation by using computer vision, image processing and pattern recognition technology. In this paper, a set of robust features are calculated from license plate characters based on angle, transit and Kirsch edge detector, and then classified using fusion of SVMs. The proposed recognition method is evaluated on a database of Iranian license plate characters consisting of 10,000 binary images, and the recognition rate of 99.68% is achieved. Also, we obtained 99.75% accuracy using four-fold cross validation technique on 10,000 dataset. Further we evaluated our method on the available dataset that contain 1200 sample. Using 70% samples for training, we tested our method on whole samples and obtained 99.85 % correct recognition rate. In addition, experimental results have demonstrated our method has better performance on Iranian license plate character recognition in comparison with contemporary methods.